I 5aA)02mo”z601 cPaper No. 205

Hypervelocity Impact of Spaced Plates by a Mock Kill Vehicle

James Wilbeck, Stephen Herwig, Jon KilpatrickITT Industries, Inc., Advanced Engineering & Sciences Division

Huntsville, Alabama

Douglas FauxLawrence Liverrnore National Laboratory

Livermore, California

Eugene S. HertelSandia National LaboratoriesAlbuquerque, New Mexico

Robert WeirSandia National LaboratoriesAlbuquerque, New Mexico

Milan Dutta,U.S. Army Space and Missile Defense Command

Huntsville, Alabama

Summary – In support of the National Missile Defense (NMD) program, a series of Light

Gas Gun (LGG) lethality tests wereconducted at the Arnold Engineering Development

Center (AEDC). Anew projectile was designed for this test series that would be

representative of aspects of a generic NMD system kill vehicle. A series of projectile

development tests were pefiormed during the design phase of the projectile. This paper

reports the results from the second development sho~ in which the projectile impacted

normally against two thin aluminum target plates, spaced approximately 5.5 diameters apart.

Results reported include the documentation of the darnage to the first and second plates, the

debris generated behind the fust plate, and correlation of these with analytical and numerical

predictions. Hydrocodes used for analyses included ALE3D, run by the Lawrence Livermore

National Laboratory, and CTH, run by the Sandia National Laboratories. The purpose of the

hydrocode analyses was to help in assessing the ability of these codes to predict the debris

formation process and the target damage.

Srmdia is a multiprogram laboratoryoperated by Sandia Corporation, aLockheed h4artin Company, for theUnited States Departnlent of Energy

1 under contract DE-AC04-94AL850W.

.:.,-f- .-inn . .. . . ,. . . . . . . . ..7..... ~., ,.. , -. . . .,., . . ., ) . .-:-.-, . . ...> , . . . . . ,,.,,, .+ ., . . . . . . . . . ..,—, ., —— ----

Paper No. 205

IMPACT CONDITIONS

This test was conducted using the large two-stage light gas gun at AEDC. This gun,

having a 14 inch (35.6 cm) diameter pump tube and a 3.3 inch (8.4 cm) launch tube, is the largest

two-stage LGG in the United States. The pump tube of this gun can be seen on the left side of

the photograph presented in Figure 1. The projectile had an overall length of 13.32 cm and an

outside diameter of 8.38 cm and was made of Lexan, Figure 2. The mass of this projectile was

held to 439 grams by drilling holes in the Lexan to lower its effective density.

TEST SET-UP

The target set-up is shown in Figure 3. The target plates were hung normal to the

shotline, and were free to swing. The first two plates were 0.25 in (0.64 cm) thick Aluminum

6061-T651, and the third plate was a 1.00 in (2.54 cm) thick steel witness plate. Due to the

limited room in the target chamber, each of the plates was limited to a width and height of 4 feet.

This was felt to be insufficient to prevent cracks from running to the edge, but was unavoidable.

The plates were spaced 18 in (45.7 cm) apart, and were held with cable in the top two corners.

Both vertical and horizontal X-rays were taken with 450 KEV X-ray heads. The X-rays cassettes

were held in trays that were permanently placed 2 feet (61 cm) off shotline. Because of the

location of the X-ray cassettes in the target chamber, the centers of the plates had to be offset a

few inches from the shotline.

TEST RESULTS

For this test, the measured impact velocity was 6.502 km/s. Laser photographs and X-

rays of the projectile in flight, examples of which can be seen in Figures 4 and 5 respectively,

showed the projectile to be in one piece at impact. The orientation of the projectile at impact of

the first plate was extrapolated from the X-rays taken in flight. These values were found to be

5.2° pitch down and 12.6° yaw right.

DISCLAIMER

This report was prepared as an account of work sponsoredby an agency of the United States Government. Neitherthe United States Government nor any agency thereof, norany of their employees, make any warranty, express orimplied, or assumes any legal liability or responsibility forthe accuracy, completeness, or usefulness of anyinformation, apparatus, product, or process disclosed, orrepresents that its use would not infringe privately ownedrights. Reference herein to any specific commercial .product, process, or service by trade name, trademark,manufacturer, or otherwise does not necessarily constituteor imply its endorsement, recommendation, or favoring bythe United States Government or any agency thereof. Theviews and opinions of authors expressed herein do notnecessarily state or reflect those of the United StatesGovernment or any agency thereof.

DISCLAIMER

Portions of this document may be illegiblein electronic image products. Images areproduced from the best available originaldocument.

..-. --. — ——,. . . .. .

.

.

Paper No. 205

As stated earlier, vertical and horizontal X-rays were taken of the debris cloud behind the

first aluminum plate, prior to its impact of the second aluminum plate. The time of the X-rays

was about 45ps after impact of the first aluminum plate. These X-rays were studied to determine

the amount of projectile erosion caused by the f~st plate and the residual projectile that would

subsequently impact the second plate. Copies of both X-rays are presented in Figure 6.

The damaged target plates were recovered and extensive measurements were made of the

two aluminum plates. A photograph of the f~st plate is presented in Figure 7. The hole in the

first plate was nearly circular, with an average diameter of 4.826 in (12.258 cm). (The chunk

cut out of the hole in the lower left side was originally assumed to be caused by a late-time

impact by a piece of gun piston, and was not included ~ the circular area.)

The pieces of the second plate were reassembled and photographed, with the results seen

in Figure 8. Petaling and extensive bending of the plate were observed. In order to quantifi the

observed damage to the plate, each section of the plate was thoroughly measured. These

measurements were then used to digitally reconstruct a flattened image of the plate. Figure 9

illustrates the results of this effort. Further quantification of darnage to the second plate was

made by studying the small holes and craters in the plate that were observed around the larger

hole in the middle of the plate. Based upon the size of the holes and depths of the craters, two

regions of damage were formulated. These regions can be seen in Figure 8, where white tape has

been used to delineate the bounds of the regions.

NIH-Image software was employed to digitally measure the area and center of the actual

hole. Next, an area-equivalent circle was calculated and overlaid on the digital image with its

center coincident to the actual hole center. This resulted in a circle centered 3.2 in (8. 13 cm) to

the right and 2.3 in (5.84 cm) below the plate center, with a radius of 7.1 in (18.03 cm). This

3

Paper No. 205

hole is outlined in red in Figure 9 and represents a reasonable axi-symmetric approximation of

the primary plate darnage.

Secondary and tertiary darnage rings were also measured and photographed. These are

outlined respectively as yellow and blue bands in Figure 9. For the purposes of this analysis,

secondary damage was defined as particle damage resulting in individual craters that fully

perforated the plate. Tertiary darnage was defined as particle damage that resulted in deep

individual craters that did not fully perforate the plate. It was observed that a reasonable

approximation of the secondary darnage radius was 9 inches, with a tertiary damage ring radius

of 13 inches. These damage bands correspond to those outlined in white tape on the actual plate

in Figure 8.

COMPARISONS WITH ANALYSES

Efforts were made to compare the data obtained from this test with various analytical and

numerical models. Two sets of comparisons will be presented in this paper. The first looks at an

analysis of the size of the hole in the first plate. The measured value was compared with a

variety of existing models and good correlation was found with two. Secondly, hydrocodes were

run by two DOE laboratories to look at the early time impact event, and predictions of damage to

the first two plates and the debris generated behind the first plate were compared with the test

results.

ANALYTICAL STUDIES

This section compares the predictions of two established hole size models against the

observed average hole diameter generated after impact with the first thin aluminum plate.

Although it is known that the impact conditions included significant projectile pitch and yaw, for

computational simplicity, none of the following models were modified to account for these

effects. All computations assumed a normal impact and utilized the following constants:

4

v

lx

PP

d

t

c

= Impact Velocity

= Target Density(6061-T6 Al) =

= Projectile Density (LEXAN) =

= Projectile Diameter (Normal Impact) =

= Target Thickness =

= Sound Speed (Al) =

Paper No. 205

= 6.502 k123/S

2.713 /#CC

1.19 g/cc

8.38 cm

0.64 cm

5.386 k123/S

One of the better-known hole-size models was developed at GM Delco in the 1960’s

~ef. 1]. Based upon experimental observations and a great deal of small iiagment data, the

model took the form

D/d = 0.45 V (tid)2’3+ 0.90 (1)

A second model developed in a similar fhshion for NASA in the 1960’s ~ef. 2] took the form

D/d = 3.4 (tid)0333(v/C)0”333 (1 -0.0308 (p@p)) (2)

Both of these models have been successfully compared with data from the hypervelocity impact

of small fragment against thin plates. The fact that they differ so greatly inform lead to concerns

as to their applicability to the reported case of a large Lexan projectile. However, both gave very

good agreement with the test. Using the values given above, the results of the predictions areas

follow:

Average Value GM Model NASA Modelfrom Test Prediction Prediction

I Predicted Hole Dia. I 12.26 cm I 11.96 cm I 11.97 cm I

The small under-prediction of diameter can easily be attributed to the pitch and yaw of the

projectile which would cause an increase in the area of projectile overlap of the target.

5

..... .x. ., . . . . . . .. .—. —..

Paper No. 205

HYDROCODE STUDIES

Hydrocode studies were conducted by both the Sandia National Laboratories (SNL) and

the Lawrence Livermore National Laboratory (LLNL). The primary focus of their analyses was

predictions of early time debris formation and resulting plate damage. Predictions of debris

behind the fwst plate were correlated with X-rays from the test. Predictions of early time plate

damage were correlated with the observed hole damage in the tested plates. Late-time structural

response of the plates was not considered. The efforts by both laboratories were quite good.

Both laboratories ran 3-D versions of their respective codes to account for projectile pitch and

yaw.

LLNL Results with ALE3D

An ALE3D simulation was performed on the second NMD development shot discussed

above. ALE3D [3] is a 3-D, arbitrary-LaGrange-Eulerian, finite element code that treats fluid

and elastic-plastic response of materials on an unstructured grid. The major components of the

code are explicit and implicit continuum-mechanics, thermal diffision, and chemistry. All

components of the code participate in advection and operate with slide surfaces. The ALE3D

code employs a relaxation scheme that allows the mesh to move into areas of high gradients and

concentrate zones in regions of interest. This methodology allows for a more accurate simulation

with fewer zones than the traditional Eulerian approach.

One reason for performing this calculation was to evaluate the equation of state used for

Lexan. The ALE3D set-up for this simulation is shown in Figure 10. The ALE3D model

contained 4.5 million elements and was run with 160 processors on the IBM SP2 massively

parallel supercomputer at LLNL. It took 58 CPU houns to run this simulations out to 150

microseconds. An initial orthogonal mesh was used and the projectile and plates where

overlayed onto the mesh. The projectile was pitched and yawed about its center of gravity (CG).

6

.

Paper No. 205

The CG was located at the Y=O plane and the Z=O plane. An initial velocity of 6.502 knds was

applied to the projectile in the positive X direction. The projectile and plates had a material

weighting factor greater than the background air resulting in the mesh relaxing (moving) into the

materials. A Mie-Gruneisen equation of state was used for the Lexan and aluminum and the

Steinberg-Guinan High Rate model was used for the constitutive response of each material.

Figure 11 is a series of material plots showing a side view QCYplane) at 25,45,65,90,

120 and 150 microseconds after impact. The projectile expands laterally at a rate of 0.12 cndps

as it traverses the space between the witness plates resulting in a debris radial spread half angle

of 22.5 degrees. Figure 12 is a 2-D material plot at 2=0 cm showing mesh location. This

illustrates the advantage of the ALE technique by concentrating zones where they are needed.

X-rays were taken at 45 ps tier projectile impact of the fnst aluminum plate. A three-

dimensional view of the projectile at 45 ps after impact is shown in Figure 13. Pseudo-

radiographs were then produced from the ALE3D dump file for comparison to experimental x-

rays. Figure 14 shows a comparison between the x-ray at about 45 ps and the ALE3D pseudo-

radiographs at 45 ps. The projectile impacted with significant yaw introducing asymmehy into

the problem which can only be captured with a three-dimensional simulation. Very good

agreement between x-rays and simulation are observed.

As discussed earlier, Figure 8 presented a photograph of the damaged second aluminum

plate. It is apparent from this figure that the impact caused extensive fracturing, generated

multiple petals, and entirely sheared off a large portion of the lower-right quadrant of this plate.

Figure 9 showed a visual reconstruction of the plate with all petals bent back flat. Figure 15

compares this sketch with the ALE3D simulation of the damage to the second aluminum plate.

To conserve problem size and code run time, only a subsection of the full plate was modeled.

The primary (red) and secondary (yellow) experimental damage rings are overlaid onto the

7

,. .

Paper No. 205

simulation showing very good agreement. The simulated center of darnage is centered 1.988 in

(5.05 cm) to the right and 0.811 in (2.06 cm) below the plate center. The second aluminum

witness plate was hung off center to make room for the x-ray film packs (1.125 in Ietl and 1.4375

in up). The corrected simulated debris hole center is 3.114 in (7.91 cm) right and 2.248 in (5.71

cm) down as compared to 3.378 in (8.58 cm) right and 2.299 in (5.84 cm) down for the

experiment. It is important to note that the ALE3D simulation does not account for fracture and

was only run to 150 microseconds, thereby not exhibiting later-time structural deformations of

the plates.

Finally, the hole size in the first aluminum witness plate is

experimental hole size, Figure 16. The average hole diameter in the first

experiment was 4.826 in (12.26 cm) and for the simulation the average

compared with

witness plate for

the

the

diameter was 4.7 in

(1 1.94 cm). It is interesting to note that the hole pattern in the first witness plate does not exhibit

fracture and petaling of the aluminum; thereby, agreeing quite well with the simulation. A

second point to note in the figure is that the ALE3D run predicts extensive plastic deformation in

the region where a tear occurred in the actual target plate. From this, it has been deduced that the

tear may have been due to the asymmetry of the initial impact, in which a small area failed due to

the excessive plastic flow. This failure caused a notch of material to rip out of the plate, with

the inertia of the tab causing the tear to be greater than the predicted plastic zone.

Sandia Results with CTH

The second NMD development shot was simulated by Sandia in two ways; the automatic

mesh refinement (AMR) option in CTH was used to simulate a two-dimensional impact of the

AEDC projectile with a spaced plate array and the standard multiprocessor version of CTH was

used to simulate the shot in three-dimensions. CTH is an Eulerian shock physics analysis

package, or hydrocode, which solves the conservation of mass, momentum and energy equations

8

.

Paper No. 205

using a finite volume scheme over a spatially fixed computational mesh. When the AMR option

in CTH is invoked, computational mesh is added and removed as the simulation progresses.

Mesh is added when material activity (material with non-zero velocity) changes. The method

uses a patch based refinement scheme, where a “patch” of additional mesh is added in a

hierarchical fashion with a 2:1 refinement ratio. In the version used for this study, patches are

added and removed depending on material activity.

CTH has models for material strength and failure, volumetric changes due to pressure and

temperature, high explosive detonation and initiation, and energy sources. It has the capability to

run on a variety of serial, distributed data parallel, and shared memory parallel computing

platforms. The user intefiace is identical for serial and parallel execution. CTH has been

validated against a large class of experimental data and is widely used in the Department of

Defense and Department of Energy laborato~ complex. Typical simulations include shock

propagation,penetration and perforation of armor, warhead mechanics, and high explosive

initiation and detonation.

The AMR option in CTH can provide much higher local mesh resolution than the

standard version of CTH due to its ability to automatically adjust the local zone size. By

weighting the local zone size with material activity in the mesh, a minimum local zone size of

0.068 cm in the 2D computational mesh (52.5 cm x 70 cm) for this problem can be provided by a

mesh size varying between 10,000 and 300,000 zones. This minimum zone size results in-9

zones through the thickness of the aluminum plates. The standard version of CTH, for the same

problem with a cell size of 0.068 cm, requires 794,766 zones in the 2D mesh, or roughly 2.5

times larger.

During this analysis, for both the standard and AMR option in CTH, the projectile was

modeled using a polycarbonate sesame EOS for the Lexan, elastic/perfectly plastic strength

9

,>,,-...-,,,~r, ,~r’-, ~-,,%> ., ,, .- .,, .s...,,........ .... ,<.>,,..,,-“ >...,,.,,.-,

Paper No. 205

model, and fracture. The aluminum plates were modeled using a pure aluminum sesame EOS,

elastic/perfectly plastic strength model, and fracture. The AMR option in CTH was run on a Sun

Ultra60 workstation for a 2D axis-symmetric representation of the impact. The case run with the

standard version of CTH was done entirely 3-dimensionally with no planes of symmetry due to

the projectile pitch and yaw and was run in parallel on the Sandia ASCI multi-processor teraflops

computer using 118 nodes.

. Figure 17 shows the 2D simulation of the debris cloud generated by the projectile impact

at 6.5 km/s using the AMR option in CTH. The simulation shows a jet forming by the portion of

the aluminum plate constrained by the projectile’s forward cup. The impact hole measures 12

cm (4.7”) in diameter and the exit hole measures 36.5 cm (14.4”) in diameter.

Figure 18 shows the standard CTH 3D simulation of the debris cloud generated by the

projectile impact at 6.5 km/s. The 3D simulation was made over a 52 cm x 52 cm by 80 cm

mesh with a uniform cell size of a 0.25 cm cube. This results in a mesh of 13,844,480 zones.

The figure shows a 2D slice through the computational space at the y=Oplane and at t=O, 30,90,

and 150 ps.

Figures 19 and 20 compare the experimental and code predicted (simulated) x-ray of the

projectile taken approximately 45 microseconds after impacting the first aluminum plate. Major

features of the projectile are replicated in the CTH calculation: overall shape and orientation, the

jet of material preceding the body of the projectile, the mushrooming of the body, and the trailing

debris on the edges. The CTH calculation shows a more pronounced span-capon the end of the

projectile. The similarities between the calculated and test projectiles at this stage after impact is

a source of confidence that CTH is modeling the material response well.

Figures 21 and 22 show the calculated holes in the first and second aluminum plates at

150 ps. The first plate has a roughly circular entry hole 12-13 cm (4.7”- 5. l“) in diameter with an

10

Paper No. 205

asymmetry reflecting the pitch and yaw of the projectile. The hole in the second plate is roughly

41 cm (16.1 in) in diameter; a ratio of roughly 4 between the two hole diameters. The similarity

in the hole sizes between the finely resolved calculation done with the 2D AMR option in CTH

and the coarser calculation done with standard 3D CTH version increases the confidence the

resolution in the 3D calculations are providing realistic simulations.

The holes in the plates that resulted from the development shot were shown previously in

Figures 7 and 8. As with the ALE3D run, CTH predicted a protrusion in the region where a

notch of material was torn out. The reported hole size is 4.8” in diameter, which compares well

to the calculated hole size for both the AMR and standard CTH simulations.

The exit hole in Figure 8 is much larger than that predicted by CTH, but this is due to the

hole shown in Figure 22 being taken at 150 microseconds. The hole will grow with time after

impact as structural response to the momentum imparted to the plate becomes dominant. A

better comparison of penetration damage is possible by comparing the prediction of Figure 22

with the digitally reconstructed second plate as shown in Fig 9. The stated size of the exit hole

is on the order of 18” in diameter, which was of the order calculated by both the AMR option

and standard CTH simulations.

CONCLUDING REMARKS

This paper reports a single test at high velocity in which the impact conditions are well

defined, the debris generated at impact are well characterized, and the damage to the target plates

are well documented. It is hoped that this data can be used by various hydrocode and structural

code modelers to validate their capabilities to predict real impact tests.

A couple of empirical hole size models developed in the 1960’s were shown to give a

very good prediction of the hole size in the fiist target plate.

11

Paper No. 205

The extensive work by LLNL using ALE3D and Sandia with CTH appears to have

demonstrated both qualitatively and quantitatively their ability to predict the complex response

of both projectile and target to the early time impact conditions. However, these hydrocode

studies also demonstrated the need for structural analysis efforts to predict the late time plate

damage.

REFERENCES

1. Maiden, C. J., Gehring, J. W., and McMillan, A. R., “Investigation of Fundamental

Mechanism of Damage to Thin Targets by Hypervelocity Projectiles,” General Motors

Corporation Final Report No. TR63-225, GM Defense Research Laboratories? Santa Barbar%

CA, September 1963.

2. Lundeberg, J. F., Stern, P. H., and Bristow, R. J., “Meteoroid Protection for Spacecraft

Structures~’ NASA Contractor Report, NASA CR-54201, October 1965.

3. Dube, E., et al, “User Manual for ALE3D An Arbitnuy LaGrange/Eulerian 3D Code

System,” February 18,2000, Version 2.9.0.

12

Paper No. 205

Figure 1 – AEDC Two-Stage Light Gas Gun Facility

Figure 2 – CAD Drawing of the Lexa.nProjectile

13

.’?rmt .>. ,, . -— . -.T-- ---- -4--- .--, . - -

Paper No. 205

X-raypreferences:1)2,.orthogonalx-rayheads2) x-ray.filmpack as closeto centerline/platesas possible3) 3 pairsOexan, alum] stepwedges loc,atedon @ach filmpack as shown4) hang projectileand take.set-upx-raysto adjustx-rayenergy5) Hoti-zontalx-rayat 35 microsecondsafterimpact6) Verticalx-ray:at65.microsecondsafterimpact

10’Diameter Shot TankHorizontalX-ray

m pack

esparallel

possible)Lookingdown range

Al AL Stl*1 *2 *3

9“

‘Ill

t.-.... ..... .—.-

X-rayCenterline

X-rayheads betweentwo AL plates

LookingEast

Figure 3- Approximate Test Setup for the Second Development Shot

14

Paper No. 205

Figure 4 – Laser Photograph of the Projectile in Flight

Figure 5 – X-ray of Projectile in Flight

Paper No. 205

Figure 6 – Horizontal and Vertical X-rays of Debris Between Aluminum Plates, T = 45ps

Figure 7 – Hole in Front Plate Target

16

..

1

7

65

Perforated Plate

Position 1

Position 3 Position 4

Position 5

Figure 8 – Photographs of the DamagedSecond Aluminum Plate

Paper No. 205

Position 6

Position 7

Paper No. 205

Figure 9 – Digitally Reconstructed View of the Second Aluminum Plate with Petals Folded In

18

Paper No. 205

L

zO

10 -

Y Rx@ -

-10 -

-20

1111111111111111$111111101!111111110 10 20 30 w Sra Sa

x Fl)cls

@oY

L x

(b)

(a)

I

x

L z

(c)

Figure 10- ALE3D set-up for Development Shot #2 simulation: a) Side view (XY plane) redprojectile, green: first aluminum plate, blue: second aluminum plate, b) Side View closeupshowing projectile pitch of 5.2° nose down, c) Top View (XZ plane) closeup showing projectileyaw of 12.6° nose right.

(a)

(c)

(e)

.. . >. -* ,. ,.”, ~’

“.

. ;.. . . .

,.

“<i:..~

$-’”’%.

... .

J ,.

i’,k

,,,.

..,

,.

6’.“d

.,.- , ;.4

.q. . .. :’. . . .-. ‘ .

(f)

Paper No. 205

(b)

(d)

“$,*,.‘,, :

., ~ .~.c .

>.. .

:.

., . .

.

‘w-* (,*

. . ... ... .:+,“ ... .

4Q..

Figure 11- ALE3D Simulation of Development Shot #2: Side Views (XY plane) at varioustimes, a) 25 b) 45, c) 65, d) 90, e) 120 and f) 150 microseconds after impact.

20

Paper No. 205

Figure 12- ALE3D Simulation of Development Shot #2: Two-dimensional slice at 45microseconds of XY plane at 2=0.0 cm. Plot illustrates the ALE technique by moving zones intoregions of interest. The Lexan (red) and front aluminum plate (green) have weighting factorsgreater than the background air.

Figure 13- ALE3D Simulation of Development Shot #2: Three-dimensional view of theprojectile and first aluminum plate at 45 microseconds. The Lexan (red) and front aluminumplate (green) are shown.

21

-,-,.:. -.,.-; -~’-~, .- r~~.------- ~ .,,.=,,., ,,. ., ,..., <. ,~,------- !,, ., . . . . . . . . . . . . -—-?-- —-- —.-.-,. -

ALE3D

Paper No. 205

. 4----==.

; ,., .

I “-%ji$ip+++

-.. s=

ALE3D

Experiment(a)

Experiment(b)

Figure 14 – Comparison of the experimental x-rays and ALE3D pseudo-radiographs of theprojectile at 45 microseconds. a) vertical x-ray plane b) horizontal x-ray plane (projectile travelsright to left)

22

Paper No. 205

Figure 15- Comparison of ALE3D simulation and experiment for the second witness plate. Topfigure is a digitally reconstructed view of the second witness plate from the experiment. Primary(red) and secondary (yellow) damage rings are shown. The experimental center of damage was3.11 inches right and 2.25 inches down versus 3.37 inches right and 2.30 inches down for thesimulation.

23

Paper No. 205

I

Figure 16- Comparison of ALE3D simulation and experiment for the first witness plate. Topfigure is a photograph of the first aluminum plate and the bottom picture is the ALE3Dsimulation showing hole size at 45 microseconds. Average hole diameter for the experiment was4.8 inches versus 4.7 inches for the simulation.

/

24

.

Paper No. 205

(c) (d)

I I

(a) (b)

Figure 17. 2D AMR option in CTH simulation of Development Shot #2: Side Views (X2 Diane).A.

at various times, a) Ob) 45, c) 100, and d) 150-microseconds after impact (domainplotted from x = –20 to x=20).

25

. ~ --rp”( mew-, .: —-.W.-/ ,.-. ~...:m?mm . .. . . ,-...-.rr , - .-,... ..-.V ..---7se .,. -.m - .. --m.- - —. -. .m. .-., . .

Paper No. 205

, ~ 2-D length y cro= wc.tion , ~ 2-D length y oross sootlon

,,

❑ ual 1Ip 1

60- ❑ m, $0

❑ YW

50 50

40 40

g 30?

30&

N 20 N 20

10

-1: T

-20

ELo;:ekl “o ‘0 3“ 50X (cm) Y=-L2491xlo-’

NlA3 HTK+I LEXAN BODY 30 Dwe.lapmenf Shot (bob)CPOBDZ 03/16/99 16W.15 CTH O Time=O.

‘F&N

(a)

,0 2-D length y crow section

~D”’

60

50

40

30

20

10

1 9,v

o——

-101

,

_~o ~

- -30 -19 10 30 so

l!kJ30, Mock 1 X (cm) Y=-L2491xlo-f

NMD HTK-8 LEXAN EOOY 30 Davdopmsnt Shot (bob)CPOBDZ 03/16/99 16W15 CTH l%’ Tlmn=9.0024M0-s

(c)

10

0

-lo1

- ❑ ml 2

❑ m

-20 ‘I

G13DI;cA 1’10 l“[cm] 30 50Y=-L249M0-’

NMTI HTK-8 LEXAN BOOY 30 Development Shot (bob)CPQBOZ 03/16/99 16X555 CTH 771 Tlme=3.0031x10-3

70

60

50

40

30

(b)

N 20

10

I

o——

-10 I

_2. ~

❑ 3D;~ck 1’10 ;’(cm] 30

50

Y=-L2491xlo-f

NMD HTK-8 LEXAN EODY 30 Owalopmnnt Shot (bob)CRRCKE 03/18/99 17:2745 CTH 3922 Tima=l.50032x10-’

(d)

Figure 18. 3D CTH simulation of Development Shot #2: Side Views (X2 plane) at varioustimes, a) Ob) 30, c) 90, and d) 150 microseconds after impact.

26

.

Paper No. 205

- ,. ,.,

- X.6Veil

1.

Figure 19- X-ray of AEDC projectileat 45 pseconds after impact

30

25

20

15

10

5

~o>

-5

-lo

-15

-20

-25

-30

‘Rate 1 ho!o”

❑- •1

- •1

m ’30’20“0 ,& ‘0 20 3030R Olock 1 Z=8.75X1O-ItWD HTK-8 lEXAtl EOOY 30 Davdapmd Shot (I+CRRCKE 05@/99 f7:265 cm 3922 Tim@=l.30032x10-4

Figure 21 – CTH Prediction of Hole inFirst Target Plate

-D radiwragh Y cro~ aooiian

(“%&

*

“n

.,--

El 3DR Black 1 Y=L2125x10iNMZ HTK-8 LEXAN BODY 341 Development Shd (bab)

COSAVU 03/15/99 l&ork26 CTH 907 Tlme=4.50762x10-5

Figure 20- Simulated x-ray of CTH Calculationat 45 pseconds after impact

.5- 0> 1-5 le h/$ ‘.. .= A41i

-30 ~

@30i:ck l-20 “o , ~) ‘0 20 30Z=&6QSx101tUO HIK-8 LSKAN 80DY Zll Oevakpmsnl Shot (bdjCRRCKE a3/13/99 IZZ65 cm 3922 TinlA.3o032xto-+

Figure 22 – CTH Prediction of Hole inSecond Target Plate

27